How to calculate what your food should have cost against what it did, read the gap, and trace it to the ingredients behind it.
By Rob Bruni, founder of OpsEdge Hospitality, twenty years running restaurants, bars, clubs and resorts. Updated November 2026.
Free printable: Food Cost Leak Audit (PDF)
Your food cost percent tells you there is a problem. Actual versus theoretical food cost tells you what kind. Theoretical food cost is what your food should have cost based on what you sold and what each dish costs to make. Actual food cost is what it did cost, from purchases and inventory. The gap between the two is food that left the building without being paid for: over-portioning, waste, comps nobody rang, theft or a bad count. This guide shows how to calculate both, how to read the gap, and how to trace it to the ingredients behind it, with a worked example.
Theoretical food cost is the sum of every item sold multiplied by its plate cost, divided by food sales. It assumes every plate was made exactly to recipe, nothing was wasted and nothing left without being rung in. It is the best food cost your menu can produce.
Actual food cost is beginning inventory plus purchases minus ending inventory, divided by food sales. It is what really happened. The food cost guide covers the formula.
The variance is actual minus theoretical, in points and in dollars. Every point of variance is food you paid for and did not sell.
Two restaurants can both run 34 percent food cost and have completely different problems:
Without the theoretical number, both owners would take the same action, and one of them would be fixing the wrong thing. That is why actual versus theoretical, often shortened to AvT, is the most useful food cost number an operator can track.
Weekly is the right rhythm. A monthly AvT finds the problem four weeks late; a weekly one finds it while the same cooks are on the same station.
One week at a full service restaurant, simplified to six menu lines:
| Item | Sold | Price | Plate cost | Sales | Theoretical cost |
|---|---|---|---|---|---|
| Grilled salmon | 310 | $29 | $9.15 | $8,990 | $2,836 |
| Ribeye | 180 | $44 | $16.40 | $7,920 | $2,952 |
| Chicken entree | 420 | $24 | $6.20 | $10,080 | $2,604 |
| Burger | 520 | $18 | $5.05 | $9,360 | $2,626 |
| Pasta | 360 | $21 | $4.35 | $7,560 | $1,566 |
| Appetizers and sides | 1150 | $11 | $2.95 | $12,650 | $3,392 |
| Total | $56,560 | $15,977 |
The menu should have produced a food cost of 28.2 percent. Here is what the inventory says it actually cost:
| Dollars | % of food sales | |
|---|---|---|
| Theoretical food cost | $15,977 | 28.2% |
| Actual food cost ($14,200 + $18,400 - $14,780) | $17,820 | 31.5% |
| Variance | $1,843 | 3.3 points |
Food cost of 31.5 percent looks like a pricing problem until you see the theoretical number. The menu is priced to run 28.2 percent; $1,843 of food left the building unpaid for this week. Over a year that is about $96,000. That is the number worth chasing.
The Actual vs Theoretical Food Cost Workbook does exactly what this guide describes: cost every recipe once, paste the weekly sales mix from your POS, enter your key item counts, and it shows theoretical food cost, actual food cost, the gap in points and dollars, and a ranked list of the ingredients behind it. Excel and Google Sheets, with a 13-week example of 26 dishes and 47 ingredients loaded.
See the AvT Food Cost Workbook Recipe Cost Card SystemFree calculator
Enter one week of numbers. It shows your actual food cost, the gap to theoretical in points and dollars, and what the gap costs over a year. The worked example above is loaded.
Theoretical food cost is each item sold times its plate cost, added up. Use the same week, cutoff and count for both.
The total gap tells you how much. Key item variance tells you where. Pick the eight to twelve items that make up most of your food spend (proteins, seafood, cheese, fryer oil) and compare what you should have used, from the recipes and the sales mix, against what you did use, from the counts and receiving.
| Key item | Should have used | Actually used | Difference | Cost of the gap |
|---|---|---|---|---|
| Salmon (lb) | 155 | 196 | +41 | $541 |
| Ribeye (lb) | 158 | 182 | +24 | $379 |
| Chicken breast (lb) | 210 | 216 | +6 | $19 |
| Ground beef (lb) | 195 | 214 | +19 | $103 |
| Fryer oil (gal) | 31 | 34 | +3 | $21 |
| Total | $1,063 |
Five items explain $1,063 of the $1,843 gap, more than half of it. Salmon and ribeye carry most of it, which points at portioning on the grill station or trim yield on the whole fish, so that is where the chef starts on Monday: weigh portions for a week, recheck the salmon yield, and watch the grill at peak. Ground beef is worth a look at the burger station; chicken and fryer oil are close enough to be counting noise.
A well-run kitchen usually lands within one to two points of theoretical, the range on our benchmarks page. No kitchen hits zero: there is always some trim, tasting and normal waste. Between two and three points, look at the key items. Three or more points is a problem that is costing real money every week, and it needs a plan with an owner.
Watch the trend as much as the number. A variance that holds at 1.5 points every week is a well-run kitchen. One that jumps from 1 to 4 in a single week usually has a single cause you can find quickly, often a count error or one big waste event.
The free food cost calculator will check the actual side in a minute.
Five free calculators turn these guides into results in a few minutes: prime cost, food cost, labor cost, break-even and pour cost. Each one shows the gap to the benchmark for your concept in dollars a year. No signup.
Open the free calculators See the benchmarksFree printable
Twenty-four places food cost leaks, from the dock to the dish pit, with an OK or FIX box for each. Walk it once a month. Two printable pages.
Here it is. We also sent a confirmation to your inbox.
Download the PDFTheoretical food cost is what your food should have cost based on what you sold and each dish's plate cost. Actual food cost is what it did cost, from beginning inventory plus purchases minus ending inventory. The difference is food that was paid for but not sold: over-portioning, waste, unrung items, theft or counting errors.
Multiply the quantity of each menu item sold by its plate cost, add the results, and divide by food sales. You need a current recipe cost for every item and a product mix report from your POS for the same period.
Within one to two points is typical for a well-run kitchen. Two to three points means it is time to look at key items. Three points or more is a real leak that needs a plan and an owner.
Weekly. A monthly comparison finds the problem four weeks late; a weekly one finds it while the same people are on the same stations. Count the key items each week and run a full inventory monthly.
The usual causes are over-portioning, waste and spoilage, comps and staff meals that were never rung in, out-of-date recipe cards or prices, receiving errors, theft and counting mistakes. Key item variance, which compares expected and actual usage of your highest-cost ingredients, shows which one it is.
